Epidemiology of Sheep Lice and Ecacy Evaluation of 60% Diazinon and 1% Ivermectin against Bovicola ovis in Sayint District, South Wollo, Ethiopia

Sheep lice, caused by Bovicola ovis are very common in Ethiopia. It can result decreased production and reproduction, downgrading and rejection of skins in tannery industries. A cross-sectional and experimental study were conducted in Sayint District, South Wollo, Ethiopia which aims to determine the prevalence of sheep lice with the associated risk factors, identify the major sheep lice species and evaluate the ecacy of 60% Diazinon and 1% Ivermectin against Bovicola ovis. Study kebeles and animals were selected using purposive and multistage sampling techniques. Accordingly, a total of 232 sheep were sampled to estimate the prevalence of sheep lice. Fifteen sheep for in-vivo and 80 Bovicola ovis lice for in-vitro test were used for drug ecacy trial, by using completely randomized design.


Conclusion
The overall prevalence of sheep lice infestation in the present study area was higher, in which Bovicola ovis was predominant and resistant against 1% Ivermectin. Therefore, tailor made intervention is required with a view to reducing the prevalence of sheep lice infestation and addressing drug resistance in the study area.

Background
Ectoparasite of sheep are very common and widely distributed in all agro-ecological zones in Ethiopia and sheep lice are considered as one of potential threats which pose a serious economic loss to the farmer, the tanning industry and the country as a whole [1,2]. Skin problems caused by lice can result in mortality, decreased production and reproduction, downgrading and rejection of skins [3]. Genus Linognatus are obligate hematophagous ectoparasites of mammals and these ectoparasites harm the health of their hosts by blood sucking and this leads primarily to enormous losses of blood. Furthermore, the biting sites become super infected with bacteria and these wounds attract licking ies [4,5]. Bovicola ovis is one of the most common lice found on sheep. They feed by chewing on the skin surface and surface debris and produce itching, irritation and possible hair loss. An allergic skin hypersensitivity reaction to lice is another cause for "Cockle" in processed sheep skins [6]. In this regard, lice infestation in sheep cause cockle and it is the commonly reported disease in Ethiopia [7][8][9].
Control of sheep ectoparasites is currently an integration of sheep husbandry, farm management and insecticide use. Chemical control remains the most important and widely used strategy against most insect pests around the world. However, studies have shown that multiple resistance mechanisms in insects confer resistance to a range of insecticide classes [10]. For instance, the result of widespread chemical use has been the sequential development of resistance of Bovicola ovis [11]. Usually, resistance will be suspected through lack of e cacy during clinical use. However, lack of e cacy could also occur due to inadequate application of a product which leads to resistant ectoparasite species against acaricides [12].
To this end, there is a paucity of information on the epidemiology of sheep lice; the risk factors associated lice infestation in the study area and insecticide e cacy in Amhara region in general and in Sayint district of south Wollo zone in particular. Therefore, this study was initiated to determine the prevalence of sheep lice in the study area, to investigate the major risk factors associated with sheep lice infestation in the district, to identify the existing sheep lice species in the study area, to evaluate the e cacy of 60% Diazinon and 1% Ivermectin against sheep lice.

Study Area
The study was conducted in Sayint district which is found in the western part of south Wollo zone, Amhara National Regional State which is located 590 km, 659 km and 189 km from Addis Ababa, Bahir Dar and Dessie, respectively. The altitude of the district ranges from 500 to 4247 m.a.s.l. The area constitutes three agro climatic zones, highland (>2500 m.a.s.l), midland (1500-2500 m.a.s.l.) and lowland (<1500 m.a.s.l.), which comprises about 42.8%, 34.6%, 22.6% of the total area, respectively [13]. The average temperature of the district ranges from 4-40 o c and annual rain fall ranges from 800 mm to 1000 mm [14]. The main rainy season in this area is between early June and the end of September, in addition a small rain occurs between early February and end of April [13]. Livestock rearing is the most important economic activity in which shoat and cattle are the major livestock species kept in the area. The total number of animals kept in the district were 141, 921 cattle, 222, 996 shoat, 23,982 equine, 143,872 poultry.

Study Animals
Local breed sheep and their crosses with Awassi of both sexes and all ages managed under extensive production system were the study animal.

Study Design
Cross-sectional study A cross-sectional study was applied from October 2018 to April 2019 in order to estimate the prevalence of sheep lice and to identify the major lice species of sheep and determine associated risk factors in the district. Sex, age, hair length, body condition, housing management (whether the sheep is housed separately or with other animals; as separate and mixed), agro ecology (highland, midland and lowland) and season of sampling (wet and dry season) was recorded and documented along with the sample collection. Sheep were categorized into two age groups; young (< one year) and adult (> one year), as described by [15]. Age estimation was done by dentition as indicated in [16]. Body condition scores of sheep were determined as either poor or good according to [17].

Sampling Method
Preliminary data were sourced from the respective district of agricultural o ce to have lists of kebeles and list of sheep producing households in the district. Purposive sampling technique was employed to select the study district based on the availability of sheep. Two stage strati ed sampling was applied and proportional samples from each kebele were drawn. Accordingly, 34 kebeles were strati ed based on agro ecology (nine highlands, nine midland and sixteen lowland). Three kebeles were selected randomly representing the three agro ecologies. Hence, 102, 60 and 70 sheep were sampled from Yegodo, Yegoda and Ada, respectively. The selected kebeles were visited and sampled twice during the study period on October 2018 and January 2019 representing the wet and dry seasons, respectively.

Sample Size estimation
The sample size calculation used in this study was calculated according to [18], using an expected prevalence of 6.94% in Kutaber [19], a 95% con dence interval and a required absolute precision of 5%. In total 99 sheep were used for this study. However, the sample size was increased to 232 to enhance the precision.

Experimental study
Due to limited number of experimental unit, completely randomized design was used to evaluate the e cacy of 60% Diazinon and 1% Ivermectin against Bovicola ovis in local sheep. About 15 naturally Bovicola ovis infested (greater or equal to 100) local male sheep aged between two to four years that have moderate body condition with short and medium hair size and not recently treated with any acaricide were used for this study. Study animals were obtained from three willing local farmers. Before the commencement of the trial, study farmers were trained how to uniformly manage their animals by the investigator and separate houses for each group of sheep was prepared and cleaned. Lists of individual sheep identi cation number were made and lottery method of randomization was used to allocate into three groups having ve sheep each, and followed for 28 days. After 15days of acclimatization period, by proper straining and handling of animal, rst group were sprayed with 60% Diazinon, the second group was given 1% Ivermectin injection subcutaneously as per manufacture's recommended dose and the third group was left untreated as a control. Both in-vivo and in-vitro e cacy trials were performed for Diazinon as per the protocols given by [20], while only in-vivo e cacy trial was conducted for Ivermectin as per [21]. Lice count was performed every 7 days for treatment and control groups and recorded starting from day 0 prior to treatment, days 7, 14, 21 and 28 according to [22]. 60% Diazinon contains an active ingredient of 600g/l Diazinon produced by Adamitulu pesticide processing share company, Addis Ababa, Ethiopia.1% Ivermectin was manufactured by Bash Pharmaceutical Company, Khartoum, Sudan, and it was imported by Rang veterinary company, Addis Ababa, Ethiopia.
In-vivo trial: Each selected sheep in treatment and control group was visually examined and lice from individual animal was counted and recorded separately for 60% Diazinon and 1% Ivermectin for the invivo evaluation. The total lice count was applied by direct examination of the body with naked eyes. Lice from each sheep was counted using parting method, classifying the body into ve parts: neck, shoulder, withers, ank and rump after marking 4 partings per site on both sides of the body [20,22]. Thus, twenty sites on each side of the body were examined by parting the eece about 10 cm and counting all live lices observed, so that , the total count from 40 sites constitutes the body count for each animal. The total lice count per animal was estimated by summation of the lice number at each site. Finally, lousicide activity was checked using arithmetic mean louse count for 60% Diazinon and 1%Ivermectin group separately along with a control group which was calculated according to [23]. Thus, mortality, 98-100% indicates susceptibility and less than 98% is suggestive of the existence of resistance [24].
In-vitro trial: su cient live and motile lice were collected manually from naturally infested sheep that came into clinics and immediately taken into the laboratory. After species identi cation, 80 Bovicola ovis louse were randomly allocated into treatment and control groups. Each group contains 4 replicates having 10 lice placed on each petri dish as per the protocol provided by [25]. Then, 60% Diazinon was diluted in water according to the manufacturer's recommendation (1:1000) and lice in the treatment group were immersed completely in 0.5ml of this solution of 60% Diazinon for one min [26] and the control lice in the same amount of distilled water. After one minute, the solution soaked and dried using Whatman lter paper. Then, the vital signs in which lice exhibited after treatment was checked at 10, 30, 60, 120 minutes, as well as 6 and 12 hours of contact time under microscope [25].
Accordingly, recording of vital signs of lice in the treatment and control groups was held in each visiting time as described by [26]. For the calculation of mortality, highly stringent criteria was used and lice only judged as dead if they are in the categories 3 or 4 after 10, 30, 60 and 120 minutes as well as after 6 and 12 hours of contact time with 60% Diazinon[26].The percentage mortality was calculated using a formula described by [27] as follows; Where the insecticidal effect of Diazinon was classi ed as: ''strong'' when mortality is >80%, ''moderate'' mortality 80-60%, '' weak'' mortality 60-40%, ''little or no activity'' mortality < 40%.

Data Collection
Clinical examination Data were collected in two seasons (October and January) to observe the seasonal lice infestation variation between the late rainy and dry seasons of the area. After proper restraining, clinical examination of each sheep was done using multiple eece partings in the opposite direction in which sheep's hair or wool normally rests. The skin was inspected carefully with naked eye and palpated across all parts of the sheep for the presence of ectoparasites and gross lesions that could be suggestive for clinical form of parasitic infestation [28]. Sheep having the parasite or lesions like alopecia and itching was considered as positive.

Lice collection and identi cation
Careful examination of the animal body, especially neck, shoulder, breast, ribs, back, and ank and rump areas was performed to detect the presence of lice by parting the body hair. Lice were collected manually and added into labelled universal bottles with 70% Ethanol. Collected samples were dispatched to Bahir Dar Animal Health Diagnostics and Investigation Center for laboratory analysis. Lice identi cation was done using stereo microscope according to the descriptions of [29].

Data Management and Analysis
Raw data obtained during sample collection and laboratory identi cation were checked, coded, and entered into Microsoft Excel spread sheet by the principal investigator and then the raw data was exported to STATA version 13 for analysis. Both univariable and multivariable logistic regression was used to examine and quantify the association between lice infestation and explanatory variables. These signi cant variables (P<0.05) by univariable logistic regression were further evaluated by multivariable logistic regression to adjust the effect of confounding. Moreover, independent sample t-test was used to compare mean lice burden between treatment and control group

Prevalence of Sheep Lice
The overall prevalence of sheep lice in the study area was 48.3% (112/232), with the highest prevalence of Bovicola ovis (83%) followed by mixed infestation of Bovicola ovis with Linognatus ovillus (17%). No sole infestation with Linognatus ovillus found by this study (Table 1). Risk factor analysis Five potential risk factors were signi cantly (P<0.05) associated with sheep lice infestation using univariate logistic regression. These included hair length, housing management, body condition, agro ecology and season (Table 2). However, these potential risk factors were further evaluated by multivariable logistic regression. According to the nal model, hair length, body condition, agro-ecology and season were signi cantly associated (P<0.05) with sheep lice infestation ( Table 2). The odds of lice infestation in medium size hair coated sheep were 2.2 times higher as compared to sheep with short hair. Likewise, the odds of lice infestation was 1.9 times higher in poorly conditioned sheep as compared to well-conditioned ones. As compared to the lowland agro-climate, the odds of sheep lice infestation were 4.4 and 3.4 times higher in midland and highland agro-climates, respectively (Table 2).

E cacy of 60% Diazinon and 1% Ivermectin
In-vitro e cacy test: based on the vital signs lice exhibit during the test, mortality was calculated. Hence, from lice treated with 60% Diazinon, 38 of them did not show any vital signs with 12 hours giving a mortality of 95%. All lice in the control group had survived during the observation period. The test acaricide has killed 2.5% of lice at 10 min, 7.5% at 30 min, and 10% at 60 min, 17.5% at 120 min, 82.5% at 360 min (6 hour) and 95% at 720 min (12 hour). There was signi cant (p=0.05) variation in percentage mortality of lice between 60% Diazinon treated and untreated group. The highest mortality was observed in the treatment group at 720 minutes (12 hour). The over all in vitro e cacy of 60% Diazinon was 95%. The in-vitro post treatment mean number of dead lice per minute of time is shown in Fig 1. The mean number of dead lice in the treated group had slightly increased from 30 to 120 min and consequently high mortality percentage was observed between 120 and 360 min, with the highest pick at 720 min Fig.   1. However, there was no any change in the untreated group throughout the study period.
In-vivo e cacy test: the overall in-vivo e cacy of Diazinon 60% was 99%. Though very slight infestation on days 7 and 14 post treatment, there was a signi cant (p<0.001) reduction in mean lice burden on treated sheep as compared to the untreated group. After the second round spray, on day 14, all sheep were found free of lice infestation while mean number of lice burden on untreated sheep was progressively increasing The mean number of lice burden reduction by 60%Diazinon among treated and untreated sheep before and after treatment is shown in Table 3. Before treatment, there was nearly similar mean lice burden on treated and untreated groups. Signi cant reduction of mean lice count was observed in treatment groups, while, a relatively high level of infestation was observed in untreated group Fig. 2. The in-vivo test showed that the e cacy of 1% Ivermectin was 81%. Before the commencement of the treatment, there was nearly similar mean lice count among treated and untreated groups. Mean lice count reduction by 1% Ivermectin injection was observed from day 0 to day 14, with a gradual increase from day 14 till the end of the experiment. A signi cant mean lice count reduction (p=0.01) was observed among treated and untreated groups ( Fig .3 Table 4).  [37](3.8%). However, this report is found lower than [38] in Aba Jima district (83%) and in Gondar [39] who reported (57%). The possible reason for such high prevalence of lice in the study area includes poor management, feed scarcity and lack of quality veterinary services. The present result is consistent with [40] in Tiyo District (49%), and [41] in Arsi Highland areas (54%).
Bovicola ovis was the dominant lice species isolated by the present study (83%) followed by mixed infestation of Bovicola ovis with Linognatus ovillus (17%). This nding is consistent with the reports of [41] and [42] who reported 86% Bovicola ovis and 12% mixed infestation (Bovicola ovis with Linognatus ovillus) in Arsi. However, it is much higher than the reports of (28) and [42] who reported 29% and 73% Bovicola ovis, respectively. Similar report was also done in Wolmera district by [43] who reported 83.23% prevalence of Bovicola ovis in sheep.
In the present study There was a signi cant (P<0.05) association between sheep lice infestation and hair length. Those sheep with medium hair coat are more likely to be infested as compared to short ones. This nding is in lined with other previous reports done by [38,34].
According to the present study, poor conditioned sheep were more likely to be infested with sheep lice as compared to well-conditioned counterparts, which agrees with the results of other studies [9,36,41,44,45] who reported poor body condition sheep were more affected by ectoparasite infestation than good body condition., Poor body condition can be exacerbated by itching and disturbance during grazing. This condition can compromise the nutritional status of sheep and leads to immune suppression. However, the present nding contradicts the ndings of [38], who reported the prevalence of Bovicola ovis was higher in well-conditioned animals.
In the present study, sheep kept in midland (OR=4.4) and highland (OR=3.4) agro-climates were more likely to be infested by sheep lice as compared to lowland agro-climates. This nding is consistent with [34,45], but it disagrees with [2,44], who reported higher prevalence of lice was recorded in Highlands followed by midland and lowland areas. The reason for higher prevalence in midland and highland might be associated with the temperature and moisture requirement of Bovicola ovis as well as shearing practice of farmers in the study area. In midland areas, temperature and moisture are conducive for lice to multiply. In highland areas, as the climate is very cold, farmers don't practice wool shearing to protect their sheep from cold stress. This practice could predispose sheep to lice infestation.
A signi cant higher lice infestation rate was observed during dry season as compared to wet season.
This nding is found compatible with other previous reports [30,46], while it disagrees with others [31,44], where higher prevalence of sheep lice was recorded in wet season than dry season. The signi cant discrepancy between dry and wet season prevalence could be associated with feed scarcity that can result in poor body condition. During dry period, feed scarcity is common which exposes sheep to starvation and poor body condition that lead to immune suppression. The overall in-vivo e cacy of 1% Ivermectin was 81%, which is below the standard for in-vivo trials set limit (98-100%), suggestive of drug resistance. This result is far lower than the ndings of [49], who reported that Ivermectin was 100% effective against sheep lice based on pour on formulations [19,47,50]. The probable reason why reduced e cacy of 1% Ivermectin injection against Bovicola ovis could be associated with prolonged and inappropriate use of the drug, the frequent application of 1% Ivermectin drug with the same brand without rotation with other insecticide classes. The importance of these presumed factors associated with drug resistance was stated by [19]. It is also evident that lack of insecticide class rotation and high treatment frequency increases macro-cyclic lactones resistance of parasites [51,52].

Conclusion
The present study revealed that there was a higher prevalence of sheep lice infestation in the study area, where Bovicola ovis was the predominant lice species affecting sheep. A number of potential risk factors were tested to determine signi cant association with sheep lice infestation. Hair length, body condition, agro ecology and season were signi cantly associated with lice infestation in sheep. The in-vivo and invitro trials showed that 60% Diazinon was more effective to treat Bovicola ovis, where as 1% Ivermectin fails to treat sheep infested with Bovicola ovis. Hence, subsequent application of tailor made intervention towards the identi ed risk factors along with rational choice of drugs against sheep lice is warranted to reduce the prevalence of sheep lice infestation in the study area. The authors declare that they have no competing interests.

Funding
No funding was obtained for this study.
Authors' contribution SL planning of the study, data collection and drafted the manuscript. MH participated in the planning of the study, revised the manuscript. HT participated in the planning of the study, analysed, interpreting and editing of the nal manuscript. YF involve during planning of the study, interpreting the nding and editing of the nal manuscript. All authors read and approved the nal manuscript. In-vitro mean dead lice count after 60% diazinon treatment in sayint district In-vivo mean lice burden of sheep treated with 60% Diazinon per different days of inspection Figure 3